NVIDIA GeForce RTX 4090 vs NVIDIA RTX PRO 6000 Blackwell Server
We compared two Desktop platform GPUs: 24GB VRAM GeForce RTX 4090 and 96GB VRAM RTX PRO 6000 Blackwell Server to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA GeForce RTX 4090 's Advantages
Lower TDP (450W vs 600W)
NVIDIA RTX PRO 6000 Blackwell Server 's Advantages
Released 2 years and 6 months late
Boost Clock has increased by 4% (2617MHz vs 2520MHz)
More VRAM (96GB vs 24GB)
Larger VRAM bandwidth (1.79TB/s vs 1.01TB/s)
7680 additional rendering cores
Score
Benchmark
FP32 (float)
GeForce RTX 4090
82.58 TFLOPS
RTX PRO 6000 Blackwell Server
+52%
126 TFLOPS
Graphics Card
Sep 2022
Release Date
Mar 2025
GeForce 40
Generation
Server Blackwell
Desktop
Type
Desktop
PCIe 4.0 x16
Bus Interface
PCIe 5.0 x16
Clock Speeds
2235 MHz
Base Clock
1590 MHz
2520 MHz
Boost Clock
2617 MHz
1313 MHz
Memory Clock
1750 MHz
Memory
24GB
Memory Size
96GB
GDDR6X
Memory Type
GDDR7
384bit
Memory Bus
512bit
1.01TB/s
Bandwidth
1.79TB/s
Render Config
-
-
-
128
SM Count
188
16384
Shading Units
24064
512
TMUs
752
176
ROPs
192
512
Tensor Cores
752
128
RT Cores
188
128 KB (per SM)
L1 Cache
128 KB (per SM)
72 MB
L2 Cache
128 MB
-
-
-
Theoretical Performance
443.5 GPixel/s
Pixel Rate
502.5 GPixel/s
1290 GTexel/s
Texture Rate
1968 GTexel/s
82.58 TFLOPS
FP16 (half)
126.0 TFLOPS
82.58 TFLOPS
FP32 (float)
126.0 TFLOPS
1290 GFLOPS
FP64 (double)
1.968 TFLOPS
Board Design
450W
TDP
600W
850 W
Suggested PSU
1000 W
1x HDMI 2.1
3x DisplayPort 1.4a
Outputs
4x DisplayPort 2.1b
1x 16-pin
Power Connectors
1x 16-pin
Graphics Processor
AD102
GPU Name
GB202
AD102-300-A1
GPU Variant
-
Ada Lovelace
Architecture
Blackwell 2.0
TSMC
Foundry
TSMC
5 nm
Process Size
5 nm
76.3 billion
Transistors
92.2 billion
609 mm²
Die Size
750 mm²
Graphics Features
12 Ultimate (12_2)
DirectX
12 Ultimate (12_2)
4.6
OpenGL
4.6
3.0
OpenCL
3.0
1.3
Vulkan
1.4
8.9
CUDA
12.0
6.8
Shader Model
6.8
